Abstract
The term acute-on-chronic liver failure (ACLF) is intended to identify patients with chronic liver disease who develop rapid deterioration of liver function and high short-term mortality after an acute insult. The two prominent definitions (European Association for the Study of the Liver [EASL] and Asian Pacific Association for the Study of the Liver [APASL]) differ, and existing literature applies to narrow patient groups. We sought to compare ACLF incidence and mortality among a diverse cohort of patients with compensated cirrhosis, using both definitions. This was a retrospective cohort study of patients with incident compensated cirrhosis in the Veterans Health Administration from 2008 to 2016. First ACLF events were identified for each definition. Incidence rates were computed as events per 1,000 person-years, and mortality was calculated at 28 and 90 days. Among 80,383 patients with cirrhosis with 3.35 years median follow-up, 783 developed EASL and APASL ACLF, 4,296 developed EASL ACLF alone, and 574 developed APASL ACLF alone. The incidence rate of APASL ACLF was 5.7 per 1,000 person-years (95% confidence interval [CI]: 5.4-6.0), and the incidence rate of EASL ACLF was 20.1 (95% CI: 19.5-20.6). The 28-day and 90-day mortalities for APASL ACLF were 41.9% and 56.1%, respectively, and were 37.6% and 50.4% for EASL ACLF. The median bilirubin level at diagnosis of EASL-alone ACLF was 2.0 mg/dL (interquartile range: 1.1-4.0). Patients with hepatitis C or nonalcoholic fatty liver disease had among the lowest ACLF incidence rates but had the highest short-term mortality. Conclusion: There is significant discordance in ACLF events by EASL and APASL criteria. The majority of patients with EASL-alone ACLF have preserved liver function, suggesting the need for more liver-specific ACLF criteria.